Mechanism of melanoma cells selective apoptosis induced by a photoactive NADPH analogue

Abstract

Melanoma is one of the most lethal cancers when it reaches a metastatic stage. Despite the spectacular achievements of targeted therapies (BRAF inhibitors) or immuno-therapies (anti-CTLA4 or anti-PD1), most patients with melanoma will need additional treatments. Here we used a photoactive NADPH analogue called NS1 to induce cell death by inhibition of NADPH oxidases NOX in melanoma cells, including melanoma cells isolated from patients. In contrast, healthy melanocytes growth was unaffected by NS1 treatment.NS1 established an early Endoplasmic Reticulum stress by the early release of calcium mediated by (a) calcium-dependent redox-sensitive ion channel(s). These events initiated autophagy and apoptosis in all tested melanoma cells independently of their mutational status. The autophagy promoted by NS1 was incomplete. The autophagic flux was blocked at late stage events, consistent with the accumulation of p62, and a close localization of LC3 with NS1 associated with NS1 inhibition of NOX1 in autophagosomes. This hypothesis of a specific incomplete autophagy and apoptosis driven by NS1 was comforted by the use of siRNAs and pharmacological inhibitors blocking different processes. This study highlights the potential therapeutic interest of NS1 inducing cell death by triggering a selective ER stress and incomplete autophagy in melanoma cells harbouring wt and BRAF mutation.

Keywords: ER stress; NADPH analogue; NADPH oxidase; ROS; melanoma.

Figure 1. NS1-induced changes in melanoma cells and melanocytes viability and in eNOS and NOX isoforms

A. Different melanoma cells or normal melanocytes were treated by 30μM NS1 or 30μM DPI or 100 μM L-NAME during 72h. The cell viability in A375 and other melanoma cells was determined by trypan blue staining. B. the percentage of ROS formed in the various melanoma cells and melanocytes (NHM) was analyzed by the measurement of fluorescence of the CellROX Deep Red reagent. DMSO: control buffer. The results are expressed as percentages of the control. C-F. Kinetics of eNOS, NOX1 and NOX4 levels in A375 melanoma cells (C) or SK-Mel 28 (D) and 1205 Lu melanoma cells (E) and melanocytes (F) were exposed at indicated times with 30μM NS1 or 30μM DPI or 100 μM L-NAME. Lysates were analyzed by western blot using the indicated antibodies. HSP90 was used as a loading control. One representative experiment out of three is shown. Data are quantified as mean ± SD of three independent experiments performed in triplicate. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 2. NS1 induces an early ER stress, modulated ROS levels and enhanced a release of calcium

A. NS1 inhibits the formation of superoxide O₂^−. in RAW 264.7 cells. Data compare the relative changes in the rates of formation of the DEPMPO-OOH spin adduct measured as the differences in the intensity of the first line of the DEPMPO-OOH spin-adduct and normalized to radical levels found in PMA-stimulated macrophages in the absence of NS1. Conditions are described in Experimental procedures and data are taken from 3-4 representative experiments. B. and C. Kinetics of ER stress markers expression in A375 melanoma cells (B) or other melanoma cells (C) were treated by 30μM NS1 or 30μM DPI or 100 μM L-NAME at indicated times. Cells were lysed and analyzed by western blotting using the indicated antibodies. HSP90 or actin was used as a loading control. One representative experiment out of three is shown. D. In parallel, calcium release was determined by monitoring the fluorescence of a calcium probe. The results are expressed as fluorescence intensity. Data are mean ± SD of three independent experiments performed in triplicate. E. A375 melanoma cells were transfected with a siRNA against Ire-1α (si Ire-1α), CHOP (siCHOP), or a siRNA control (siCtrl). Twenty-four hours after transfection, cells were treated with 30μM NS1 or DMSO for 72 hours. The results are expressed as percentages of the control. In parallel, cells were lysed and analyzed by western blotting using the indicated antibodies. HSP90 was used as a loading control. The results are expressed as percentages of the control.

Figure 3. Effect of NS1 on autophagy: A375 melanoma cells

A. or other melanoma cells B. were treated by 30μM NS1 or 30μM DPI or 100 μM L-NAME at indicated time. Cells were lysed and analyzed by western blotting using the indicated antibodies. HSP90 was used as a loading control. One representative experiment of three is shown. C. Immunofluorescence pictures of A375 melanoma cells treated with DSMO or 30μM NS1. LC3 was labeled with a secondary red antibody, NS1 was visualized after an excitation at 480nm (green) and DNA was visualized with DAPI (blue). Note the close localization between NS1 and LC3 seen by yellow/orange puncta clearly seen in the enlarged merged view.

Figure 4. Evaluation of the autophagy in response to NS1 treatment using siRNAs and drugs

A. A375 cells were treated with E64d + pepstatin (10μg/ml each) after 72 hours, and their viability was estimated by trypan blue staining. Analysis of different factors by WB is shown below. B. A375 cells were treated with chloroquine (100 μM) after 24 hours, and evaluated as in A. C. A375 melanoma cells were transfected with a siRNA against ATG5 (siATG5) or a siRNA control (siCtrl), treated as in A. D. A375 melanoma cells were transfected with a siRNA against Ire-1α (siIre1α) or a siRNA control (siCtrl), treated as in A. E. A375 melanoma cells were treated with a proteasome inhibitor Mg132 (10 μM) after 16 hours, and their viability estimated by trypan blue staining. The short treatment duration with chloroquine and Mg132 avoided their known cell toxicity after long exposure. In parallel, cells were lysed and analyzed by western blotting using the indicated antibodies. HSP90 was used as a loading control. One representative experiment of three is shown. The results are expressed as percentages of the control. Data are mean ± SD of three independent experiments performed in triplicate.

Figure 5. Evidence for apoptosis of melanoma cells treated by 30 μM NS1

A375 melanoma cells A-B. or other melanoma cells C, D. were treated by 30μM NS1 or 30μM DPI or 100 μM L-NAME or H2O2 at the indicated times. Cells were detached and their DNA contents were measured by flow cytometry; representative raw data of annexin V, DAPI and a double labeling are shown in panel A and B. In parallel, lysates were analyzed by western blot using the indicated antibodies (A lower part) and C). HSP90 was used as a loading control. E. A375 melanoma cells were transfected with a siRNA against caspase 3 (siCasp3) or a siRNA control (siCtrl), treated as in Figure 2E. The results are expressed as percentages of the control. Data are mean ± SD of three independent experiments performed in triplicate. One representative experiment out of three is shown.

Figure 6. Effect of NS1 on signalling pathways

A375 melanoma cells A. or other melanoma cells B. were treated by 30μM NS1 or 30μM DPI or 100 μM L-NAME at indicated time. Cells were lysed and analyzed by western blotting using the indicated antibodies. HSP90 was used as a loading control. One representative experiment of three is shown.

Figure 7. Scheme of potential mechanisms for NS1-induced melanoma cell death

By a direct effect of NS1 on a calcium redox-dependent ion channel or by an indirect effect of NS1 mediated by eNOS inhibition in close proximity to caveolae, NS1 induced an early ER stress arising from a release of calcium sensed at the ER by Ire-1α, calnexin, BIP and PERK that triggered JNK, p38 and NF-kB signalling. NS1 inhibition of angiogenesis and cell cycle blockade via NOX4 decrease could be associated with down-regulation of STAT3 signalling. These redox, calcium and metabolic stresses triggered autophagy, evidenced by an increased formation of LC3-II in autophagosomes that co-localized with NS1. NS1 inhibition of ROS formed by NOX1 likely blocked autophagy relying on ROS for further fusion of autophagosomes with lysosomes required for autophagy completion. This partial autophagy is consistent with the build-up of p62 selective autophagy cargo, accumulation of DNA damage and specific effects of siRNAs on the autophagy and cell survival. This autophagy blockade resulted in cell death by apoptosis in melanoma cells carrying or not different mutational statuses.